farcasterunion

I made a graph with the ratings of my Caster Credibility Score (CCS). However, there are limitations to this approach because of some penalties in my system, and because it favors current active casters, and not casters who were active in the past.

So, the graph only shows spam label 2.
- Current active users came mostly from Feb. - Jun. 2024 (no surprise there)
- Current not or barely active users (<5 casts in 365 days) came also mostly from Feb. - Jun. 2024
- We must be careful to interpret the growth overtime, because it only shows in with time period the current FIDs in a specific rating were active in the past.

I'm still going to think about how I can do this better. CCS is more about the current value, and not about value in the past, and that makes it problematic to use in this kind of analyzes.

Interests: crypto, economics, TA & blockchain related innovations | 🔗 https://linktr.ee/cryptonjal 🔗 https://dune.com/nhejyht

Thanks.
The script is publicly available: https://dune.com/queries/5497120
So, people can fork it, analyze it, etc. 
If you put your name or FID in the search bar, you will find your own score. If you fork it, you can unhide some columns to analyze your subscores even more (hidden to make the list more readable).
The script will be materialized every week to a dataset: dune.nhejyht.result_caster_credibility_score_ccs

So you can go in all directions with it.

Applied AI, databases, data, vibe coding (sic.), open source, community/devrel, remote. 

itinerant businessperson. globalist.

Kuala Lumpur, Federal Territory of Kuala Lumpur, Malaysia

Thanks! Why do you have a score of 74.1, me 72.8, and how do we aim to increase that? I'm clearly being a bit lazy in asking you! elvi gets 98.1! so i guess it also has a lot to do with engagement per cast

much coolness, can we get our own CCS ratings? :)

101 $tipn